Fix out-of-bound issue in call to LUT in GEM-CSC trigger #30564
Conversation
|
The code-checks are being triggered in jenkins. |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-30564/16801
|
|
A new Pull Request was created by @dildick (Sven Dildick) for master. It involves the following packages: L1Trigger/CSCTriggerPrimitives @cmsbuild, @rekovic, @benkrikler can you please review it and eventually sign? Thanks. cms-bot commands are listed here |
| return std::make_pair((getLUT()->get_csc_wg_to_gem_roll(theParity))[alct.getKeyWG()].first, | ||
| (getLUT()->get_csc_wg_to_gem_roll(theParity))[alct.getKeyWG()].second); | ||
| const auto& mymap(getLUT()->get_csc_wg_to_gem_roll(theParity)); | ||
| return std::make_pair(mymap.at(alct.getKeyWG()).first, mymap.at(alct.getKeyWG()).second); |
There was a problem hiding this comment.
isn't this the same as
| return std::make_pair(mymap.at(alct.getKeyWG()).first, mymap.at(alct.getKeyWG()).second); | |
| return mymap.at(alct.getKeyWG()); |
?
|
The code-checks are being triggered in jenkins. |
|
+code-checks Logs: https://cmssdt.cern.ch/SDT/code-checks/cms-sw-PR-30564/16854
|
|
Pull request #30564 was updated. @cmsbuild, @rekovic, @benkrikler can you please check and sign again. |
|
please test |
|
The tests are being triggered in jenkins.
|
|
urgent |
|
+1 |
|
Comparison job queued. |
|
merge |
|
Comparison is ready Comparison Summary:
|
dildick
deleted the
from-CMSSW_11_2_X_2020-07-06-1100-gem-csc-trigger-bugfix
branch
PR description:
This PR should fix a bug reported in the GEM-CSC trigger in CMSSW_11_0_X samples, when a pad number goes out of bounds in the
[]operator. I also replaced[]withat()and added a few assert statements. The bug was introduced when the pad numbering in GE1/1 was changed from 1->192 to 0->191, and in GE2/1 from 1->384 to 0->383.PR validation:
Tested on 4000 events in Run-3 sample
/DYToLL_M-50_TuneCP5_14TeV-pythia8/Run3Winter20DRMiniAOD-DRFlatPU30to80_110X_mcRun3_2021_realistic_v6-v2/GEN-SIM-RAW. Also tested with WF 23234.0if this PR is a backport please specify the original PR and why you need to backport that PR:
N/A
FYI @eyigitba @tahuang1991